Positioning text in digital designs based on an underlying image

Abstract

The present disclosure is directed toward systems and methods for determining the optimal placement, color, and size of an overlay in a digital design. For example, in response to determining that a change to an underlying image of a digital design has caused a loss of readability and/or aesthetic value in the digital design due to the placement of the overlay over the now changed underlying image, one or more embodiments described herein utilize a scoring system to identify optimal placement, color, and size for the overlay in relation to the underlying image in the digital design.

What is claimed is: 
     
       1. In a digital medium environment for editing digital images, a method for automatically repositioning overlays in digital designs upon changing of underlying images in the digital designs comprising:
 detecting a replacement of a source file of an underlying image of a digital design with a source file of a new underlying image, the digital design comprising an overlay including text; 
 performing edge detection, in response to the detected replacement of the source file of the underlying image of the digital design with the source file of the new underlying image, on the new underlying image to determine a location of one or more visual edges in the new underlying image; 
 determining, by at least one processor, a placement score for each of a plurality of potential placements of the overlay relative to the new underlying image based on an amount that the overlay overlaps the one or more visual edges in each of the potential placements; and 
 repositioning the overlay over the new underlying image at a potential placement with a highest placement score. 
 
     
     
       2. The method as recited in  claim 1 , further comprising detecting a change between one or more of a color palette of the underlying image and a color palette of the new underlying image, a size of the underlying image and a size of the new underlying image, or a format of the underlying image and a format of the new underlying image. 
     
     
       3. The method as recited in  claim 1 , further comprising, in response to the detected replacement of the source file of the underlying image of the digital design with the source file of the new underlying image, facial recognition on the new underlying image. 
     
     
       4. The method as recited in  claim 3 , wherein determining the placement score for each of the plurality of potential placements of the overlay relative to the new underlying image is further based on an amount that the overlay overlaps one or more identified faces. 
     
     
       5. The method as recited in  claim 4 , further comprising:
 identifying one or more colors associated with the overlay, and one or more colors associated with a portion of the new underlying image that falls underneath the potential placement with the highest placement score; 
 determining whether the one or more colors associated with the overlay adversely affects readability of the text of the overlay; and 
 changing, based on the determination, the one or more colors associated with the overlay to increase the readability of the text. 
 
     
     
       6. The method as recited in  claim 5 , further comprising:
 identifying a size associated with the overlay; 
 determining whether the size associated with the overlay adversely affects readability of text; and 
 changing, based on the determination, the size associated with the overlay to increase the readability of the text. 
 
     
     
       7. The method as recited in  claim 4 , wherein determining, by at least one processor, a placement score for each of a plurality of potential placements of the overlay relative to the new underlying image based on an amount that the overlay overlaps the one or more visual edges in each of the potential placements further comprises:
 calculating a plurality of scalar values, each scalar value indicating an effect of positioning the overlay in a given potential placement; and 
 summing the plurality of scalar values for each potential placement. 
 
     
     
       8. The method as recited in  claim 7 , wherein calculating a plurality of scalar values comprises calculating an edge scalar value indicating how the overlay overlaps the one or more visual edges of the new underlying image when positioned in a given potential placement. 
     
     
       9. The method as recited in  claim 7 , wherein calculating a plurality of scalar values further comprises calculating a face scalar value indicating how the overlay overlaps with faces of the new underlying image when positioned in a given potential placement. 
     
     
       10. The method as recited in  claim 7 , wherein calculating a plurality of scalar values further comprises calculating a symmetry scalar value indicating how the overlay, when positioned in a given potential placement, affects the symmetry of the digital design. 
     
     
       11. The method as recited in  claim 7 , wherein calculating a plurality of scalar values further comprises calculating an alignment scalar value indicating how the overlay, when positioned in a given potential placement, aligns with objects in the digital design. 
     
     
       12. A non-transitory computer-readable medium storing instructions thereon that, when executed by at least one processor, cause a computer system to:
 detect a replacement of a source file of an underlying image of a digital design with a source file of a new underlying image, the digital design comprising an overlay; 
 perform edge detection, in response to the detected replacement of the source file of the underlying image of the digital design with the source file of the new underlying image, on the new underlying image to determine a plurality of potential placements in the digital design for text relative to one or more visual edges in the new underlying image of the digital design; 
 determine a placement score for each of the plurality of potential placements that indicates an effect of positioning the text in each of the potential placements relative to one or more visual edges in the new underlying image of the digital design; 
 place the text over the new underlying image at a potential placement with a highest calculated placement score; 
 extract a color palette from the new underlying image; and 
 modify one or more colors of the text to correspond to the extracted color palette. 
 
     
     
       13. The non-transitory computer-readable medium as recited in  claim 12 , wherein the instructions, when executed by the at least one processor, cause the computer system to perform edge detection to determine a plurality of potential placements by dividing the digital design into a grid. 
     
     
       14. The non-transitory computer-readable medium as recited in  claim 12 , further storing instructions thereon that, when executed by the at least one processor, cause the computer system to identify one or more features of the digital design. 
     
     
       15. The non-transitory computer-readable medium as recited in  claim 14 , wherein the instructions, when executed by the by the at least one processor, cause the computer system to identify the one or more features of the digital design by performing one or more of facial recognition, symmetry detection, size detection, alignment detection, or overlap detection. 
     
     
       16. The non-transitory computer-readable medium as recited in  claim 15 , wherein the instructions, when executed by the by the at least one processor, cause the computer system to determine the placement score for each of the potential placements by calculating the placement score based on one or more of the performed facial recognition, symmetry detection, size detection, alignment detection, or overlap detection. 
     
     
       17. The non-transitory computer-readable medium as recited in  claim 12 , further storing instructions thereon that, when executed by the at least one processor, cause the computer system to:
 determine whether the one or more colors of the text adversely affects readability of the text; and 
 modify the one or more colors of the text further by changing an opacity, shade, or hue of the one or more colors to increase readability of the text. 
 
     
     
       18. The non-transitory computer-readable medium as recited in  claim 12 , further storing instructions thereon that, when executed by the at least one processor, cause the computer system to:
 determine whether a size of the text adversely affects readability of the text; and 
 change, based on the determination, the size of the text to increase readability of the text. 
 
     
     
       19. A system comprising:
 at least one processor; and 
 at least one non-transitory computer-readable storage medium storing instructions thereon that, when executed by the at least one processor, cause the system to: 
 detect a replacement of a source file of an underlying image of a digital design with a source file of a new underlying image, the digital design comprising an overlay including text; 
 perform edge detection, in response to the detected replacement of the source file of the underlying image of the digital design with the source file of the new underlying image, on the new underlying image to determine a location of one or more visual edges in the new underlying image; 
 determine a placement score for each of a plurality of potential placements relative to the new underlying image based on an amount that the overlay overlaps the one or more visual edges in each of the potential placements; 
 reposition the overlay over the new underlying image at a potential placement with a highest placement score; 
 modify one or more colors of the repositioned overlay based on an extracted color palette from the new underlying image; and 
 modify a size of the repositioned overlay based on readability of the repositioned overlay. 
 
     
     
       20. The system as recited in  claim 19 , wherein determining a placement score for each of the plurality of potential placements relative to the new underlying image further comprises identifying one or more features of the digital design by performing one or more of facial recognition, symmetry detection, size detection, alignment detection, or overlap detection.